Hi Jordi > From: On Behalf > Of Jordi Gutiérrez Hermoso > Sent: 24 August 2011 21:48 > To: Kris Thielemans > > On 24 August 2011 15:35, Kris Thielemans > <kris.f.thielem...@gmail.com> wrote: > > [suggestions to improve mkoctfile] > > I had similar thoughts not long ago. See this thread: > > > <http://octave.1599824.n4.nabble.com/I-think-I-quot-broke-quot-mkoctfile-for -3-4-1-td3608378.html> >
ah. sorry. Seems that my search missed this. Apologies one comment on jwe's reply: jwe> should setting CFLAGS="" on the jwe> command line append or replace the default CFLAGS? Maybe we should jwe> have CFLAGS="..." replace and CFLAGS+="..." append? given the matlab situation, I'd vote for CFLAGS="..." to replace. CFLAGS+="..." to append seems to be a nice idea. Supporting matlab's syntax of doing CFLAGS="$CFLAGS ..." to append seems also good (but it does brings us into nasty quoting/escaping territory). > In particular, notice what jwe said at the end. There are in effect > *two* mkoctfiles, one in shell scripts, and the other in C++ > sources for the benefit of Windows users. > > The proper fix is probably thus to abandon the shell version > and instead only fix the C++ version, because we really don't > want to be maintaining two scripts. right. I was wondering on how this happened on plain windows. (I'm using cygwin so am effectively using a linux build of octave.) I agree that a single version of mkoctave is best, but don't have enough experience with building octave to vote for the script or the C++ version. Thanks Kris ------------------------------------------------------------------------------ EMC VNX: the world's simplest storage, starting under $10K The only unified storage solution that offers unified management Up to 160% more powerful than alternatives and 25% more efficient. Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev
